Design and Implementation of a Monopulse Sonar Transmitting Array Submitted To Kyungtae Han Dr. Brian L. Evans Professor, The University of Texas at Austin Dr. Terry L. Henderson Research Scientist, Applied Research Laboratories Prepared By Matthew Riley EE464 Senior Design Project Electrical and Computer Engineering Department University of Texas at Austin Spring 2005 CONTENTS LIST OF TABLES..............................................................................................................iv LIST OF FIGURES.............................................................................................................v EXECUTIVE SUMMARY.................................................................................................vi 1.0 INTRODUCTION.......................................................................................................1 2.0 DESIGN PROBLEM STATEMENT........................................................................1 3.0 DESIGN PROBLEM SOLUTION............................................................................2 3.1 MONOPULSE SONAR PROCESSING..........................................................2 3.1.1 Traditional Monopulse.............................................................................2 3.1.2 Transmit-Monopulse Theory...................................................................4 3.2 SONAR TRANSDUCER ARRAY DESIGN...................................................5 3.3 ECONOMIC ANALYSIS..................................................................................7 4.0 DESIGN IMPLEMENTATION................................................................................7 4.1 MATLAB SIMULATION.................................................................................7 4.1.1 Transmitter Simulation............................................................................8 4.1.2 Receiver Simulation................................................................................10 4.2 PROTOTYPE CONSTRUCTION.................................................................10 5.0 TEST AND EVALUATION.....................................................................................11 5.1 MATLAB SIMULATION...............................................................................11 5.2 IN-WATER EXPERIMENTS........................................................................13 5.2.1 Experimental Setup.................................................................................13 5.2.2 Experimental Results..............................................................................14 6.0 TIME AND COST CONSIDERATIONS...............................................................16 7.0 SAFETY AND ETHICAL ASPECTS OF DESIGN..............................................17 8.0 CONCLUSIONS AND RECOMMENDATIONS..................................................17 REFERENCES...................................................................................................................18 APPENDIX A – SIMULATION SOFTWARE SOURCE CODE...............................A-1 APPENDIX B – PROTOTYPE SCHEMATICS...........................................................B-1 ii CONTENTS (Continued) APPENDIX C – EXPERIMENT SUPPORT SOFTWARE SOURCE CODE..........C-1 i ii LIST OF TABLES 1 Monopulse shading function element weights.................................................................8 2 Optimal simulation system parameters..........................................................................12 2 RMS error for most accurate in-water experiments.......................................................15 i v LIST OF FIGURES 1 Monopulse receiver system design..................................................................................3 2 Monopulse transmitter system design..............................................................................4 3 Transducer array layout and numbering system..............................................................6 4 Monopulse shading functions..........................................................................................9 5 Simulation results for varying SNR...............................................................................12 6 Experimental setup flow diagram..................................................................................14 7 Most accurate in-water experiment results....................................................................15 v EXECUTIVE SUMMARY Monopulse sonar processing is an important branch of sonar research that explores methods for calculating the arrival angles of underwater sound waves. This paper documents my senior design project, which explored the implementation of monopulse processing techniques in a sonar transmitter. The application of monopulse technology in the transmitter rather than the receiver, referred to as “transmit-monopulse,” is a new extension of existing monopulse theory that has not, until now, been explored by the research community. I begin by explaining the motivations for pursuing transmit- monopulse technology, and define the requirements of a monopulse sonar transmitter design for a number of real-world applications. I continue by summarizing basic existing monopulse techniques, and then showing how these techniques can be applied in transmit- monopulse. I then illustrate how I implemented a design for a monopulse transmitter in both software and in a functional prototype. I later discuss the results predicted by my software simulation and the implications of those predictions on the actual data obtained through in-water experimentation. Finally, I recommend ways in which future research can extend the ideas developed in my project. In addition to presenting an itemization and summary of the design and implementation process, I discuss challenges that arose over the course of the project. Issues like time and economic constraints, as well as difficulties in prototype construction caused an eventual need to reformulate my project plans. In this paper I briefly explain how these issues affected the final outcome of the project. At the end of the paper I discuss the ethical and safety concerns involved in my project, and provide a list of references used to research monopulse technology. Appendices A, B, and C include source code for my software simulation, prototype schematics, and the source code for software I developed to automate experiments. v i 1.0 INTRODUCTION This report comprehensively documents the design and implementation of a monopulse sonar transmitting array. In the work presented here I discuss my design process, detail my design’s implementation, and analyze the performance of my final design solution. I performed this project as an undergraduate student employee of the Applied Research Laboratories, University of Texas at Austin (ARL-UT). Monopulse sonar processing research is an exciting branch of work performed at ARL, and the monopulse transmitter I designed is a novel application of monopulse theory developed at the lab. Dr. T. L. Henderson developed the bulk of the theory implemented in my project, and it was under his supervision that my work was performed. The work presented here is a natural extension of existing monopulse sonar research, and with its completion, the potential arises for many new applications of monopulse sonar technology. I begin my report by discussing the design problem that my project addresses, and I illustrate example applications of my final design. I then develop basic monopulse processing principles and describe the application and implementation of those ideas in my design solution. Finally, I present the results of a series of in-water experiments conducted with a prototype of my design. I assess the validity of these results with respect to theoretical values, and conclude by recommending future extensions of my work. 2.0 DESIGN PROBLEM STATEMENT Sonar applications such as target tracking and underwater surveying have motivated the development of a number of techniques for making high-resolution calculations of the arrival angle of an incoming sound wave. When employing most of these techniques, the system’s resolution is determined by the beamwidth achieved by the system’s transducer array. A small beamwidth allows for higher resolution measurements, and the beamwidth is inversely proportional to the physical size of the array. For my project, I wanted to develop a system that was small enough to lend itself to numerous applications while maintaining a high enough resolution to remain useful. Because my design required a small physical array dimension, which in turn limited the system’s resolution, I chose to research and implement monopulse processing techniques, which allow angle measurements to be made at higher resolution than the system’s beamwidth [1]. Existing research on monopulse techniques concentrates on systems in which the burden of monopulse processing performance must be placed on the sonar system’s receiver—a restriction that severely limits the application of this technology. The limitations of this type of monopulse processing can be seen, for example, when trying to land a small underwater vehicle on the back of a submarine. In this situation, the vehicle’s operator requires an accurate measurement of the angular offset of his vehicle with respect to the submarine. A standard monopulse receiver implementation would require extensive processing hardware to be placed in the small vehicle. In my project I devise a method for performing monopulse processing in the sonar system’s transmitter rather than in the receiver, which allows simple receivers to make high-resolution angle measurements. The application described here and those similar to it will benefit greatly from a monopulse transmitter implementation. 3.0 DESIGN PROBLEM SOLUTION 3.1 MONOPULSE SONAR PROCESSING Because my final solution stems directly from traditional monopulse theory, I will present the basics of a monopulse receiver implementation before discussing the method used to implement monopulse processing in a transmitter. 3.1.1 Traditional Monopulse Receiver Implementation In a receiver implementation, monopulse techniques dictate that the receiver simultaneously process the recorded sounds of a linear hydrophone array twice - the first time with a typical array element weighting that minimizes sidelobes, and the second time 2 with an element weighting that is the derivative of the first. A unique property of these derivative matched weightings is that the arrival angle of an incident plane wave of sound can be calculated simply by taking the ratio of the two sets of outputs [2, p.183]. To prepare the outputs for the ratio calculation, a filter must process the outputs of each weighting function. In order to resolve the derivative-matched relationship of the monopulse weighting functions, the filter for the first monopulse function output must be the derivative of the filter used for the second monopulse function output. Figure 1 depicts a model of a monopulse sonar receiver implementation. Figure 1. Monopulse receiver system diagram. When a discrete-element receiver array is employed, the angle indicated by the ratio calculation will actually be an “inflected” angle measurement, proven in [3, p.45] to be: ⎡⎛ D⎞ ⎤ tan⎢⎜π ⎟sinθ(t)⎥ ~ ⎣⎝ λ⎠ ⎦ s(t) = (1) D π λ where θ(t) is the correct angle measurement. For cases in which the source signal is a simple sinusoid, the inflected measurement can be corrected by simply solving for θ(t): 3 ⎛ ⎡~ ⎛ D⎞⎤⎞ ⎜arctan⎢s(t)⎜π ⎟⎥⎟ ⎜ ⎣ ⎝ λ⎠⎦⎟ θ(t) = arcsin (2) ⎜ D ⎟ ⎜ π ⎟ ⎜ λ ⎟ ⎝ ⎠ 3.1.2 Transmit-Monopulse Theory In a transmit-monopulse system, the objective is to transmit a signal using a transducer array such that a single hydrophone receiver with no knowledge of its location in the water can determine its angular offset relative to the transmitting array by recording and processing the transmitted sound wave. Further, a transmit-monopulse system must allow multiple receivers at different locations in the water to record the same transmission and still determine each of their respective angular offsets. The monopulse receiver is an excellent starting point for devising a monopulse transmitter. Figure 2 illustrates the final diagram of my system, which I will reference as I develop the ideas behind monopulse transmitter design. Figure 2. Monopulse transmitter system diagram. In sonar systems, the processes of transmitting and receiving can be reversed and the resulting system output will be symmetric. For example, if a given receiving transducer array with a certain shading function produces a maximum output for a 100 kHz sine wave produced at a 20-degree offset in front of it, then that same transducer array transmitting a 4